Automation and the Future of Data Science

From the moment robots first appeared on the factory floor, experts have questioned their impact on the future of the workforce. While the first industrial revolutions – steam, electricity, and digital – created job losses on farms, the assembly line and in call centers, they also created new categories of jobs. After all, someone must design, build and maintain the robots.

Today, artificial intelligence and machine learning are poised to alter the cubicle landscape, affecting white collar jobs from accounting and law to data processing and, yes, even data science.

In fact, experts believe the next round of automation almost certainly will redefine responsibilities in most occupations. Some jobs will disappear, of course, but others are more likely to become partially automated. According to McKinsey Global Institute, 60% of all occupations have at least 30% of their activities that can be automated. But these tasks tend to be highly structured, repetitive and routine.

That’s why highly skilled technology workers, such as data scientists and analysts, are more likely to benefit from automation, MGI says. Demand will rise for professionals who can design and maintain the technologies that are driving this revolution, such as virtual robots, artificial intelligence, and machine learning.

The tasks with the highest potential for automation are data processing and data collection, MGI says. In addition, Gartner believes 40% of data science tasks could be automated by 2020. These include data integration and model building. Automation can finish these tasks faster and reduce the risk of error, but it also frees up time for data scientists to work on more complex algorithms.

To derive insights and solve problems, you need humans.

Indeed, in data science and analytics, there are limits to what automation can do. To derive insights and solve problems, you need humans. That’s why demand for data scientists and analysts remains high. MGI predicts there will be a shortage of up to 250,000 data scientists in the U.S. alone over the next decade.

And for the second year, Glassdoor named data scientist as the best job in America. Yet the exact combination of skills – coding, statistics, machine learning, database management, visualization techniques, and industry-specific knowledge – is so hard to find that data scientists have earned the name “unicorns.” It’s an almost impossible combination to find.

Data science encompasses high-level tasks that can’t be automated easily.

At IQ Workforce, we don’t expect to see demand drop for data scientists anytime soon. For one thing, the role requires many high-level tasks that can’t be automated easily. They still require humans to perform. For example, data wrangling requires good human judgment, and executives will still need an articulate data scientist to walk them through the data, as well as interpret data and create visualizations that drive home the most important points.

But we also believe that automation and artificial intelligence will increase demand. Although IBM is pitching its Watson natural language analytics platform as a way to replace or supplement data scientists, it’s more likely automation will enable data science to scale. This, in turn, will create the need for talent that can work with artificial intelligence and automation.

Our clients are asking for talent that not only excels in programming and mathematics, but also is creative, collaborative, innovative, and has strong business acumen. It’s a tall order – and the key reason why there is such strong demand for “unicorns.”

So, if you’re looking for a data science opportunity – and you meet this definition – give us a call. We’ll always talk to a unicorn.